The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute the ani CLI tool to perform searches and fetch data.
Evidence: Instructions in SKILL.md detail the construction of shell commands like ani search "<query>" and ani info <id> --json based on user input.
Context: This is the core purpose of the skill. Command execution is restricted to the specific ani tool provided by the author.
[EXTERNAL_DOWNLOADS]: The skill requires an external binary to be installed on the user's system.
Evidence: README.md and SKILL.md point to https://github.com/pavelsimo/ani for installation via go install.
Context: The resource belongs to the skill's author ('pavelsimo') and is necessary for the skill's stated functionality.
[PROMPT_INJECTION]: The skill ingest untrusted content from the AniList API (such as synopsis text and user-generated recommendations) which could theoretically contain malicious instructions.
Ingestion points: Data enters the agent's context through the output of ani commands, specifically in the info and search modes.
Boundary markers: None explicitly defined in the prompts, though the agent is instructed to format output as Markdown tables and lists.
Capability inventory: The agent has the ability to execute shell commands (ani) and write Markdown responses.
Sanitization: The skill instructions recommend truncating synopsis text to 300 characters, which provides a minor reduction in the attack surface, but no explicit sanitization of Markdown or control characters is mandated.

search-anime